Global data assimilation systems for numerical weather prediction are characterized by significant uncertainties in tropical analysis fields. Furthermore, the largest spread of global ensemble forecasts in short range on all scales is in the tropics. The paper to appear in Monthly Weather Review suggests that these properties hold even in the perfect-model framework and the ensemble Kalman filter data assimilation with a globally homogeneous network of wind and temperature profiles. Results obtained by applying the modal decomposition to analysis and forecast uncertainties provide the scale-dependent efficiency of data assimilation to reduce the prior uncertainties in the balanced and inertio-gravity dynamics.  Figure shows the zonally-averaged time-averaged spread of 12-hour forecasts of specific humidity in the global EnKF system in autumn 2008.